Claude Skills by yanacuti1121

Resilience patterns for AI agent external calls. Circuit breakers, exponential backoff with jitter, stream recovery, batch error handling, and log fallback under disk pressure. Sources: nodeshift/opossum, softonic/axios-retry, vimeo/player.js, trentm/node-bunyan, visionmedia/batch.

Design resilient distributed services — circuit breaker, retry with backoff, timeout, bulkhead isolation, rate limiting, and fallback strategies. Use when asked about "circuit breaker", "retry logic", "timeouts", "cascading failures", "service keeps going down", "bulkhead", "rate limit", or "what happens when a dependency is slow/down". Do NOT use for: SLO target design — use `slo-design`.

'Reverse engineers malicious Android APK files using JADX decompiler

'Reverse engineers .NET malware using dnSpy decompiler and debugger to

'Reverse engineers iOS applications using Frida dynamic instrumentation

'Reverse engineers malware binaries using NSA''s Ghidra disassembler

Reverse engineer ransomware encryption routines to identify cryptographic

Reverse engineer Rust-compiled malware using IDA Pro and Ghidra with

Claude Code plugin tự động thiết kế agent team + sinh skill từ domain description — 6 architecture patterns, tạo .claude/agents/ + .claude/skills/ tự động. +2098⭐/week.

ReWOO decoupled planning — Planner/Worker/Solver split. 5x fewer tokens than ReAct on HotpotQA, +4% accuracy. Plan-and-Execute generalization, planner distillation to 7B. When to use plan-first vs interleaved. Sources: rohitg00/ai-engineering-from-scratch (Apache-2.0).

Curriculum AI engineering từ toán học lên production — 503 bài học, 20 phase, 320 giờ. MIT license. Build từ first principles: backprop, transformer, LLM, agent, swarm.

LiteParse — Rust document parser của LlamaIndex: PDF/Word/Excel/PPT/image → JSON với bounding boxes. Node.js/Python/WASM bindings. Tesseract OCR bundled, không cần cloud.

Runtime operational metrics with meters, timers, histograms, and moving averages. node-measured patterns for tracking request rates, execution durations, error rates, and EWMAs for agent performance observability. Sources: caustik/node-measured.

Low-level OCI container runtime patterns using Linux namespaces and cgroups. Namespace isolation (pid/net/mnt/uts/ipc), cgroup resource caps, rootless execution, OCI bundle structure, and runc lifecycle (create/start/kill/delete). Sources: opencontainers/runc.

Strict HTML cleanup with self-closing tag repair, attribute allowlists, URL protocol validation, and CSS sanitization. apostrophecms/sanitize-html patterns for cleaning LLM-generated and web-scraped HTML. Sources: apostrophecms/sanitize-html.

AI job search system trên Claude Code — A-F scoring 10 dimensions, 15 slash commands, Playwright portal scanning 45+ companies, ATS-optimized PDF resume generation.

Scan container images for known vulnerabilities using Anchore Grype with

'This skill covers integrating Aqua Security''s Trivy scanner into CI/CD

Trivy is a comprehensive open-source vulnerability scanner by Aqua Security

Tenable Nessus is the industry-leading vulnerability scanner used to

Perform security risk analysis on Kubernetes resource manifests using

'Performs advanced network reconnaissance using Nmap''s scripting engine,

Direct PubMed and NCBI E-utilities search workflows for biomedical literature, MeSH queries, PMID lookup, citation retrieval, and API-backed literature monitoring.

Store, access, rotate, and audit secrets safely — provider selection (Vault, AWS Secrets Manager, GCP Secret Manager), runtime injection, rotation policy, least-privilege IAM, CI/CD secret handling, and secret scanning in pipelines. Use when asked to "store API keys", "rotate secrets", "Vault", "AWS Secrets Manager", "secret rotation", "least privilege", "secrets in CI", "secret scanning", "gitleaks", "trufflehog", "don't hardcode credentials", or "how to manage secrets in production". Do NOT...

Secure secrets management for agent runtimes. One-time-token gateway, AES-256-GCM at-rest encryption, Shamir's Secret Sharing for API keys, memory zeroization after use, and PBKDF2 key derivation.

'Securing API Gateway endpoints with AWS WAF by configuring managed rule

'This skill guides practitioners through hardening AWS Identity and Access

'Securing AWS Lambda execution roles by implementing least-privilege

'This skill instructs security practitioners on deploying Microsoft Defender

'Securing container registry images by implementing vulnerability scanning

Harbor is an open-source container registry that provides security features

'This skill covers hardening GitHub Actions workflows against supply

Secure Helm chart deployments by validating chart integrity, scanning

'This skill covers hardening and securing process historian servers (OSIsoft

'This skill covers hardening managed Kubernetes clusters on EKS, AKS,

'This skill covers implementing secure remote access to OT/ICS environments

'This skill covers security hardening for serverless compute platforms

보안 파이프라인 - CWE Top 25 + STRIDE 자동 검증

Self-Refine iterative improvement (generate/feedback/refine loop, +20 avg across 7 tasks) and CRITIC external verification (tool-grounded critique for factual tasks). Anthropic evaluator-optimizer pattern. Sources: rohitg00/ai-engineering-from-scratch (Apache-2.0).

Mine session history for recurring mistakes and distill them into short CLAUDE.md rules. Use when asked to 'rút rule', 'tự rút rule', 'extract rules from sessions', 'lỗi lặp đi lặp lại', 'review past sessions for mistakes', 'optimize CLAUDE.md from history', 'gom lỗi thành rule', or 'what do I keep fixing'. Do NOT use for: generating a CLAUDE.md from scratch — see claude-md-enhancer. Do NOT use for: one-off retrospectives — see session-wrap.

Use when asked about citation counts, paper impact, who cited a paper, research influence, finding references of a paper, author h-index, or publication metrics. Triggers on: 'citation count', 'who cited', 'how many citations', 'paper impact', 'semantic scholar', 'research influence', 'find references', 'author publications', 'h-index', 'số lần trích dẫn', 'tìm tài liệu tham khảo', 'bài báo có ảnh hưởng'.

Build and optimize serverless functions — AWS Lambda cold starts, Cloudflare Workers edge patterns, function composition, idempotency, dead-letter queues, concurrency limits, observability, and IaC deployment with SAM/SST. Use when asked about "Lambda", "serverless function", "cold start", "Cloudflare Workers", "edge function", "AWS SAM", "SST framework", "function timeout", "Lambda concurrency", "DLQ", "dead-letter queue", "idempotent Lambda", "serverless observability", or "deploy serverles...

Load development context at session start: git branch, uncommitted changes, CONTEXT.md / TODO.md / open issues. Use when: starting a new coding session, resuming interrupted work, or when the agent needs to orient itself quickly before acting. Inspired by: disler/claude-code-hooks-mastery session_start pattern.

Auto-extract a reusable skill from the current session conversation. Use when the user says "save this as a skill", "turn this into a skill", "I want to reuse this workflow", "make this repeatable", "skillify this", "we just did something I want to automate", or after completing a useful multi-step task that could be repeated. Do NOT use for one-off fixes or tasks with no repeatable value.

세션 종료 전 자동 정리 스킬. 4개 병렬 subagent가 문서 업데이트, 반복 패턴, 학습 포인트, 후속 작업을 동시 탐지하고, 1개 검증 subagent가 중복 제거 후 사용자에게 선택지를 제시한다. 트리거: /session-wrap, 세션 마무리, 세션 정리, 작업 마무리

Configure agent-tail log aggregation for the current project. Auto-detects framework (Vite, Next.js, plain Node, monorepo) and sets up CLI runner, browser log plugins, and output destinations. Use when setting up agent-tail, configuring dev server logging, or piping logs for AI agent consumption.

sGLang RadixAttention — KV cache stored in a radix tree, reused across requests sharing common prefixes. Cache-aware scheduling (depth-first, LRU at branch level). 29% throughput edge vs vLLM on ShareGPT, 6.4x on RAG workloads, 86% hit rate on voice. Deployed 400k+ GPUs 2026. Sources: rohitg00/ai-engineering-from-scratch (Apache-2.0).

Build and customize UI components with shadcn/ui — component installation, CSS variable theming, Radix UI primitive composition, variant patterns with cva(), dark mode wiring, and extending components without forking. Use when asked about "shadcn", "shadcn/ui", "shadcn component", "radix primitive", "cva variants", "shadcn theme", "extend shadcn", "shadcn dark mode", "shadcn form", "shadcn dialog", "shadcn button variants", or "how to customize shadcn components". Do NOT use for: full design ...

100+ AI Agent & RAG app templates sẵn sàng chạy — 14 categories: starter agents, multi-agent teams, voice AI, MCP agents, RAG, memory, fine-tuning. Apache-2.0.

Analyze session work and automatically convert reusable patterns into Claude Code skills. Use when: "세션을 스킬로", "스킬 만들어", "이거 스킬로", "skill factory", "이 작업 자동화해", "스킬 추출", "make this a skill", "extract skill", "convert to skill", "스킬 팩토리", "자동 스킬 생성". Differs from skill-creator (archived) and manage-skills (drift detection): this skill actively analyzes sessions, checks for duplicates, and creates skills via Agent Teams.